• International Journal of Computer Science & Network Security
• /
• v.23 no.5
• /
• pp.53-64
• /
• 2023

Early detection of endometrial carcinoma in uterus is essential for effective treatment. Endometrial carcinoma is the worst kind of endometrium cancer among the others since it is considerably more likely to affect the additional parts of the body if not detected and treated early. Non-invasive medical computer vision, also known as medical image processing, is becoming increasingly essential in the clinical diagnosis of various diseases. Such techniques provide a tool for automatic image processing, allowing for an accurate and timely assessment of the lesion. One of the most difficult aspects of developing an effective automatic categorization system is the absence of huge datasets. Using image processing and deep learning, this article presented an artificial endometrium cancer diagnosis system. The processes in this study include gathering a dermoscopy images from the database, preprocessing, segmentation using hybrid Fuzzy C-Means (FCM) and optimizing the weights using the Whale Optimization Algorithm (WOA). The characteristics of the damaged endometrium cells are retrieved using the feature extraction approach after the Magnetic Resonance pictures have been segmented. The collected characteristics are classified using a deep learning-based methodology called Long Short-Term Memory (LSTM) and Bi-directional LSTM classifiers. After using the publicly accessible data set, suggested classifiers obtain an accuracy of 97% and segmentation accuracy of 93%.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.12
• /
• pp.4971-4975
• /
• 2015

Background: Several methods have been presented for the evaluation of the endometrium in patients with abnormal uterine bleeding, which include minimal invasive and invasive approaches such as diagnostic curettage or endometrial biopsy by Pipelle. Many studies have been performed in order to compare two methods; diagnostic curettage and outpatient endometrial biopsy. This investigation compared sampling adequacy, endometrial histopathology, failure rates, duration and costs between diagnostic curettage in a hospital and endometrial biopsy. Materials and Methods: This single blind clinical trial was performed on 130 patients older than 35 years who was referred to Amir training hospital in 2013 for elective diagnostic curettage because of abnormal uterine bleeding. For all patients eligible for the study, an endometrial sample by Pipelle was taken without anesthesia or dilatation. Then under general anesthesia diagnostic curettage was performed by sharp curette. Sampling duration was calculated and both samples were sent to the same pathologist. The diagnostic values of two methods in the diagnosis of normal endometrium, endometrial hyperplasia and carcinoma were compared. The costs of these two methods were also compared. Data analysis was performed by SPSS (version 16.0) software. Chi-Square, Fisher, and Pearson tests were used and were considered statistically significant at P values less than 0.05. Results: Two methods were agreed upon 88% of sampling adequacy and 94% of pathological results. Specificity of 100% and sensitivity of 90% for detection of proliferative endometrium, secretory endometrium, simple hyperplasia without atypia and 100% for cancer were recorded. Pipelle diagnostic accuracy in comparison with curettage, have been reported over 97%, so the failure rate in this study was below 5%. Sensitivity of Pipelle for detection of atrophic endometrium was reported below 50%. Duration and cost was lower in Pipelle versus curettage. Conclusions: It is concluded that due to high agreement and cohesion coefficient between curettage and Pipelle on the issue of sampling adequacy, histopathology finding (except atrophic endometrium), low failure rate, duration of sampling and cost, Pipelle can be introduced as a suitable alternative of diagnostic curettage.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.12
• /
• pp.7635-7638
• /
• 2013

Background: Sperm-associated antigen 9 (SPAG9) has been recently proposed as a novel biomarker for early diagnosis of several human tumors, including ovarian, cervical and breast cancers. Its clinical value remains to be clarified for endometrial cancer (EC). In this study, we investigated the utility of serum SPAG9 levels in diagnosis of EC and its association with important clinicopathological parameters. Materials and Methods: This cross-sectional study was performed at a tertiary women's referral center in Ankara, Turkey. Preoperative serum samples were collected from patients surgically treated for endometrial cancer between June 2012-April 2013. Similar aged women with a biopsy proven benign endometrium were used as controls. Serum SPAG9 levels were measured with an enzyme-linked immunosorbent assay (ELISA) method and assessed for links with clinicopathological factors. Receiver operating characteristic (ROC) curve analysis was performed to assess power of SPAG9 levels for EC prediction. P values less than 0.05 were considered statistically significant. Results: A total of 63 women with EC and 27 with benign endometrium were included in the study. Mean age in the EC group was $58.7{\pm}1.1$. Median SPAG9 levels in the EC and control groups were 18.3 (range, 12.7-53.8) and 14.1 (range, 4.3-65.3), respectively (p<0.001). A cut-off value of 17 ng/ml for SPAG9 predicted presence of malignant endometrium with 74% sensitivity and 83% specificity [Area under curve (AUC)=0.82, p<0.001]. SPAG9 levels did not demonstrate any significant association with histological type, FIGO stage, tumor grade, size, myometrial invasion, lymphovascular space invasion, cervical involvement, adnexal involvement, peritoneal cytology or lymph node status (all p>0.05). Conclusions: Testing for SPAG9 may be useful for early detection of EC in asymptomatic high-risk women. Its role in post-treatment follow-up and early detection of recurrence should be assessed in future trials.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.2
• /
• pp.769-773
• /
• 2013

Background: A literature review on 1,104,269 cancer patients concluded that the prevalence of multiple primary malignancies (MPM) is between 0.73% and 11.7%. MPMs seem to have higher incidence than that influenced by hazard only. The purpose of this study was to investigate clinically useful information for effective screening for synchronous and metachronous second primary cancers and to identify a potential surveillance protocol. Materials and Methods: Using statistical and epidemiological indicators we evaluated the patients with MPMs (double locations) admitted to Dr. Abdurrahman Yurtarslan Ankara Oncology Education and Research Hospital between 1981 and 2010. Results: Out of the 130 cases, 24 (18.4%) were synchronous while 106 cases (81.6%) were metachronous tumours. Mean interval time from first to second primary cancers was 4.65 years (0-27 years). The most frequent malignant associations were breast-breast, breast-endometrium and breast-ovary. Both primary and secondary tumors tended to be in an advanced stage explained by the low compliance of the patients to follow-up. Conclusions: The possibility that MPMs exist must always be considered during pretreatment evaluation. Screening procedures are especially useful for the early detection of associated tumors, whereas careful monitoring of patients treated for primary cancer and a good communication between patients and medical care teams should ensure early detection of secondary tumors, and subsequent appropriate management.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.13
• /
• pp.5355-5358
• /
• 2014

Background: To assess the role of sonographic endometrial thickness and hysteroscopic polyp size in predicting premalignant and malignant polyps in postmenopausal women. Materials and Methods: A total of 328 postmenopausal women with abnormal uterine bleeding and thickened endometrium underwent operative hysteroscopy due to detection of endometrial polyps were included in this retrospective study. Preoperative endometrial thickness measured by transvaginal ultrasonography and polyp size on hysteroscopy were noted. Hysteroscopic resection with histology was performed for endometrial polyps. Endometrial thickness and polyp size were evaluated on the basis of final diagnosis established by histologic examination. Receiver operator characteristic curves were calculated to assess the sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of endometrial thickness and polyp size for detecting pemalignant and malignant polyps. Results: Premalignant and malignant polyps were identified in 26 (7.9%) of cases. Sonographic measurement showed a greater endometrial thickness in cases of premalignant and malignant polyps when compared to benign polyps. On surgical hysteroscopy, premalignant and malignant polyps were also larger. Endometrial thickness demonstrated a sensitivity of 53.8%, specificity of 85.8%, PPV of 24.6% and NPV of 95.6% at a cut-off limit of 11.5 mm with diagnostic accuracy of 83.2%. Polyp size has a diagnostic accuracy of 94.8% with a sensitivity of 92.3%, specificity of 95.0%, PPV of 61.5% and NPV of 99.3% at a cut-off point of 19.5mm. Conclusions: Endometrial thickness measured by transvaginal ultrasonography is not sufficient in predicting premalignant and malignant endometrial polyps in postmenopausal women with abnormal uterine bleeding and thickened endometrium. Polyp size on hysteroscopy is a more accurate parameter, because of better sensitivity and specificity. However, while polyp size ${\geq}19.5mm$ seems to have a great accuracy for predicting premalignancy and malignancy, histologic evaluation is still necessary to exclude premalignant and malignant polyps.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.11
• /
• pp.4705-4709
• /
• 2014

Objective: To determine the accuracy of magnetic resonance imaging (MRI) in detection of metastasis in pelvic and para-aortic lymph nodes from different gynecological malignancies. Materials and Methods: This retrospective cross sectional analytic study was conducted at the Department of Diagnostic Radiology, Aga Khan University Hospital Karachi Pakistan from January 2011 to December 2012. A sample of 48 women, age range between 20-79 years, fulfilling inclusion criteria were included. All patients had histopathologically proven gynecological malignancies in the cervix, endometrium or ovary and presented for a pretreatment MRI to our radiology department. Results: MRI was 100% sensitive and had a 100% positive predictive value to detect lymph node metastasis in lymph nodes with spiculated margins and 100% sensitive with a 75% positive predictive value to detect lymph node metastasis in a lymph node with lobulated margins. The sensitivity and positive predictive value of MRI to detect heterogeneous nodal enhancement were 100% and 75% respectively. Conclusions: Our study results reinforce that MRI should be used as a modality of choice in the pretreatment assessment of lymph nodes in proven gynaecological malignancies in order to determine the line of patientmanagement, distinguishing surgical from non-surgical cases.